Hypothesis testing 1 Hypothesis 2 Graph Ho mu 20 Null H mu not equal 20 Alternative Put conditions to prove here ex less than mu 1 tail if greater than equal to less than equal to 2 tail both sides find Zcrit or Tcrit with tables 3 Decision rule 4 Calculate test stat If test stat z is 1 96 or 1 96 reject Ho Otherwise do not reject 5 Conclusion Reject Ho Do not reject Ho 6 Evidence supports the claim that Evidence does not support the claim Pop std deviation known Sample std deviation known Pop Proportion t x s n z p p p 1 p n z x n Point estimate p x n 2 population means Pop std deviation known Sample std deviation known Pop Proportion z x 1 x 2 1 1 2 2 n1 n2 Point estimator of P when p1 p2 p3 p n1p1 n2p2 n1 n2 t x 1 x 2 s1 1 s2 2 n1 n2 DF n1 n2 2 Point estimate p1 x n Standard Error p 1 p 1 n1 1 n2 Point estimate p2 x n Test Stat p 2 z p 1 p 1 p 1 n1 1 n2 Z values 05 01 1 tail 1 65 2 33 2 tail 1 96 2 58 F value chart DFTR across lower DFE down level of significance k number of groups treatments n sample size Ho mu1 mu2 m3 H At least one is different Sum of squares SS Given SSTR Degrees of freedom DF k 1 DFTR Given SSE n k DFE Treatment s Error Mean Square F value MSTR SSTR DFTR MSE SSE DFE MSTR MSE Total K 3 N 47 Treatment s Error Total SST Sum of squares SS 1200 n 1 Degrees of freedom DF 3 1 2 600 1800 47 3 44 n 1 Mean Square F value MSTR 1200 600 2 MSE 600 13 64 44 600 43 99 13 64
View Full Document